Indian Institute of Management Calcutta. ATR Adaptive Communications. Birla Institute of Technology. Joka, Diamond Harbour Road. Research Laboratories.
Improving System Performance of Ad Hoc Wireless Network with Directional Antenna S. Bandyopadhyay, M.N. Pal, Dola Saha
Tetsuro Ueda, Kazuo Hasuike
Ranjan Pal
Indian Institute of Management Calcutta Joka, Diamond Harbour Road Calcutta 700104 INDIA
ATR Adaptive Communications Research Laboratories Seika-cho, Kyoto 619-0288, JAPAN
Birla Institute of Technology Meshra, Ranchi 835215 INDIA
Abstract— It has been shown that use of directional antenna in the context of ad hoc wireless networks can largely reduce radio interference, thereby improving the utilization of wireless medium. However, that alone does not always guarantee improvement in overall system performance. In this paper, we have identified several criteria and investigated their interrelationships and impact on overall system performance in this context. Our methodology uses optimisation techniques using multicriteria decision analysis. We use analytic hierarchy process (AHP) to identify relative weights of different criteria under different application-specific scenario in order to solve the optimisation problem for each scenario by TOPSIS approach. The result shows that the parameter setting required to get optimum performance is application-specific; depending on the situation or application-scenario, several parameters need to be controlled to get better system performance.
I.
INTRODUCTION
In ad hoc wireless networks [1], all the nodes are usually equipped with omni-directional antenna. However, ad hoc networks with omni-directional antenna uses omni-directional RTS/CTS based medium access mechanism that wastes a large portion of the network capacity by reserving the wireless media over a large area. As a result, lot of nodes in the neighborhood of transmitter and receiver has to sit idle, waiting for the data communication between transmitter node and receiver node to finish. To overcome this problem, researchers have proposed to use directional or adaptive antenna that would largely reduce radio interference, thereby improving the utilization of wireless medium and consequently the network throughput [2-5]. However, in the context of ad hoc networks, it is difficult to find ways to control the direction of such antenna for transmission and reception in each terminal in order to achieve an effective multi-hop communication between any source and destination. In other words, in order to fully exploit the capability of directional antenna, whenever a source S and destination D engage in a communication, all the neighbors of source and destination nodes should know the direction of communication so that they can initiate new communication in other directions. In [2], an adaptive MAC protocol has been proposed, where each node keeps certain neighborhood information dynamically through the maintenance of an Angle-SINR Table in order that each node knows the direction
of communication events going on in its neighborhood at that instant of time. However, periodic exchange of this information increases the control overhead of the system. If the system is highly mobile, then each node has to update this angle-specific neighborhood information more frequently, which would increase the control overhead further. Additionally, if the transmission beam-width of a transmitter node towards a receiver is narrow, the receiver may quickly move out of the transmission zone so formed and the communication would be interrupted. Thus, even if we design effective MAC and routing protocols with directional antenna, that may not guarantee overall improvement in system performance. For example, what should be the optimum value of transmission-beam-angle α and transmission range R for better system performance? Will that be application-scenario-specific or it would be the same for every scenario? Assume that it is possible to get a directional antenna that can form a very narrow beam. Will that be more effective in every context? In this paper, we have identified several criteria and investigated their inter-relationships and impact on overall system performance in this context. Our methodology uses optimisation techniques using multicriteria decision analysis. We use analytic hierarchy process (AHP) to identify relative weights of different criteria under different applicationspecific scenario in order to solve the optimisation problem for each scenario by TOPSIS approach. The results show that the parameter setting to get optimum performance is applicationspecific; depending on the situation or application-scenario, several criteria need to be controlled for better system performance. II. SYSTEM DESCRIPTION An ad hoc network consists of N nodes distributed over a twodimensional space of area A. Let us assume that each node n is equipped with a directional antenna whose beam-angle can vary from 30 degree to 360 degree. Each node has a transmission range R and is moving with a velocity v units/sec. within the specified area A. Neighbors of n: Neighbors of n∈N is a set of nodes within the omni-directional transmission range R of n. α ): When a node n forms a transmission transmission_zonen (α beam with a beam-angle α (30°
